Accurate estimation of body composition in middle-aged and elderly individuals is limited by a lack of knowledge of the fat free body (FFB) composition in this age group. Currently available techniques are indirect and usually based on FFB composition of young adults. When applied to other age groups a constant FFB composition in all groups must be assumed. However, reductions in bone mineral content and body potassium with increasing age are well documented, with concommitant and significant reductions in FFB density. The limited available evidence suggests body water also declines with age. Consequently, the assumption of constant FFB composition in different groups is invalid for the older population, and errors of estimation of percent body fat as high as five percent can occur when methods developed on young adults are applied to the older population. Hence, the purpose of the proposed research will be to quantify the fat free body composition of sedentary and physically active men and women so that more accurate estimates of body composition can be developed for these age groups for such methods as densitometry, hydrometry, anthropometry, and total body electrical impedance. Two hundred and forty men and women, aged 40-80 years, will be tested over three years. Both sedentary and active subjects will be studied to test the hypothesis that activity history, in addition to age and gender, is an important determinant of body composition. All subjects will be measured for total body water, bone mineral content, body density, selected skinfolds and anthropometric dimensions, and total body electrical impedance. Bone mineral content will be measured using single and dual photon absorptiometry with measurements made at four sites on the forearm, spine and femur giving a better estimate of mineral than has previously been obtained. The subjects will be grouped by age, gender and physical activity level, and the data analyzed for group differences. Multiple regression prediction equations will be developed for estimation of fat and fat free body emphasizing a multiple component approach with body water, mineral content and body density as important components. Also, with knowledge of the fat free body water and mineral content, fat free body density may be estimated, making possible the development of improved two component equations for estimation of body fat from body density and the skinfold-body density relationship in the middle-aged and older populations. The results will have important clinical and field application in diverse areas concerned with nutritional assessment and epidemiological investigation of health and fitness, as well as research studies investigating body composition and functional capacity of middle-aged and older men and women.